AI Deepfake CSAM Class Action Expands, Adds Stability AI
A class action lawsuit against Elon Musk’s SpaceXAI (formerly xAI) and Stability AI over the creation and distribution of deepfake child sexual abuse material has expanded significantly, with two new plaintiffs joining the case and Stability AI added as a defendant. The amended complaint, filed July 7 in the U.S. District Court for the Northern District of California, alleges that the companies’ AI models were used to generate thousands of sexually explicit images of minors and that SpaceXAI obstructed law enforcement investigations into the abuse.
Background of the Case
The lawsuit, originally filed in March 2026 by three teenage girls from Tennessee, has grown to include five plaintiffs — all identified as Jane Does — from four states. The case targets SpaceXAI’s Grok chatbot and Stability AI’s Stable Diffusion image generator, alleging that both companies designed and released AI tools capable of producing child sexual abuse material (CSAM) without adequate safety guardrails.
According to NPR, the suit accuses the companies of producing CSAM, benefiting from sex trafficking ventures, negligence, defective product design, and creating a public nuisance. The plaintiffs are seeking monetary compensation and injunctive relief requiring the installation of more effective guardrails.
The Jane Doe 4 Case: A Wyoming Nightmare
Perhaps the most harrowing account comes from Jane Doe 4, a woman in her 20s from Wyoming. Her stepfather used SpaceXAI’s Grok chatbot to generate approximately 7,000 sexually explicit images and videos from a single photograph taken when she was about 11 years old. The AI-generated content included depictions of her nude, performing sexual acts on men — including her stepfather — with explicit captions.
SpaceXAI sent only one tip regarding Jane Doe 4 to the National Center for Missing and Exploited Children (NCMEC) in February 2026 — when the stepfather asked the model to generate an image depicting the girl being raped by multiple men. The company did not include any of the abusive images with its report and failed to share the alleged perpetrator’s IP address with NCMEC and law enforcement, even after officials requested it multiple times, according to the complaint.
Law enforcement eventually tracked down the stepfather through other means. Two days after being charged with child exploitation offenses, he died by suicide. As Ars Technica reported, law enforcement told Jane Doe 4 that her stepfather had used Grok because it was more responsive to his prompts than other AI models.
“A person I trusted took a photo of me as a little girl sleeping on the couch, wearing an oversized panda pajama shirt, and Grok turned that photo into thousands of sexually explicit images of me — images so horrific I can’t even begin to describe them,” Jane Doe 4 said in a statement. “When law enforcement came looking for the perpetrator, xAI went silent.”
The Jane Doe 5 Case
Jane Doe 5, a minor from Wisconsin, was 14 years old in a photograph taken at her eighth-grade graduation. An adult male relative of one of her classmates used the photo to generate CSAM through Grok and distributed it to other predators online. The perpetrator was arrested and charged with extensive possession and distribution of CSAM content.
Stability AI Added as Defendant
The amended complaint also names Stability AI as a new defendant in connection with claims filed by the three original Tennessee plaintiffs. The lawsuit alleges that Stability AI’s open-weight models were trained on CSAM and serve as the basis for third-party “nudify” apps that Grok users rely on.
A June 2026 research paper found that the Stable Diffusion family is the primary driver of image-based nudification online, accounting for 42.7% of such images. According to the Lieff Cabraser law firm, Stability AI had previously filtered NSFW content from Stable Diffusion Version 2 in November 2022 but rolled back those guardrails after user backlash over censorship.
“When they realized that nobody wanted to use their model when they put those guardrails in place, they rolled those guardrails right back so that everybody would want to use their model again,” said Annika K. Martin, lead counsel for all five plaintiffs. “They knew exactly how to rein this behavior in and they chose not to for profit.”
Stability AI responded by saying, “Any suggestion that safety is not a top priority for us is categorically wrong. We take our ethical responsibilities seriously.”
Reporting Failures and Systemic Issues
By early 2026, NCMEC found that 90% of xAI’s CyberTipline reports were not actionable by law enforcement because xAI declined to include user information. NCMEC reported more than 1.5 million CyberTipline reports in 2025 indicating a nexus to generative AI and child sexual exploitation. In more than 133,000 cases, NCMEC lacked sufficient information to determine how the technology was used.
Broader Implications and What’s Next
The lawsuit seeks class-action status on behalf of two nationwide classes — one against SpaceXAI and one against Stability AI — covering all U.S. persons whose real images as minors were altered using the companies’ tools to produce sexually explicit content. Lawyers estimate thousands of minors may be eligible to join.
Annika K. Martin described the situation as a “scourge on society” that touches every community. “The scale of harm is staggering, and the companies whose products enable it must be held accountable,” she said.
The case raises urgent questions about AI accountability, the adequacy of existing legal frameworks, and whether companies can be held liable for designing products that enable the creation of exploitative content. As the litigation progresses, it is likely to shape the legal boundaries of AI liability for years to come.